[CodeStyle][Typos][A-[21-30]] Fix typos (allctor,almostly,alog,Aread,Allways,alway,ane,adn,expaned,annotaions,annotaion,architecure,architecuture,aer)

_typos.toml

@@ -10,7 +10,7 @@ grad = "grad"
 arange = "arange"
 ot = 'ot'
 pash = 'pash'
-
+anc = 'anc'
 # These words need to be fixed
 ontext = 'ontext'
 Direcly = 'Direcly'
@@ -65,7 +65,6 @@ bootom = 'bootom'
 Vetical = 'Vetical'
 mantain = 'mantain'
 patition = 'patition'
-almostly = 'almostly'
 Dowloading = 'Dowloading'
 Prepar = 'Prepar'
 precent = 'precent'
@@ -117,7 +116,6 @@ decalared = 'decalared'
 coalesc = 'coalesc'
 graident = 'graident'
 infered = 'infered'
-Allways = 'Allways'
 substitue = 'substitue'
 Ouput = 'Ouput'
 witk = 'witk'
@@ -127,7 +125,6 @@ staticly = 'staticly'
 emited = 'emited'
 repalce = 'repalce'
 GARD = 'GARD'
-annotaions = 'annotaions'
 gloabl = 'gloabl'
 devide = 'devide'
 zerp = 'zerp'
@@ -155,7 +152,6 @@ endianess = 'endianess'
 VAILD = 'VAILD'
 ues = 'ues'
 algorithem = 'algorithem'
-aer = 'aer'
 elemenents = 'elemenents'
 CANN = 'CANN'
 pathes = 'pathes'
@@ -226,8 +222,6 @@ Rto = 'Rto'
 tunning = 'tunning'
 kerenl = 'kerenl'
 Temperarily = 'Temperarily'
-alway = 'alway'
-ane = 'ane'
 cliping = 'cliping'
 DEIVCE = 'DEIVCE'
 neeed = 'neeed'
@@ -334,7 +328,6 @@ effeciently = 'effeciently'
 workround = 'workround'
 fucntion = 'fucntion'
 sturcture = 'sturcture'
-branchs = 'branchs'
 udpated = 'udpated'
 vunerability = 'vunerability'
 funtion = 'funtion'
@@ -535,7 +528,6 @@ neigbhors = 'neigbhors'
 subsitute = 'subsitute'
 futher = 'futher'
 vart = 'vart'
-architecure = 'architecure'
 passs = 'passs'
 Propogation = 'Propogation'
 partion = 'partion'
@@ -558,7 +550,6 @@ outputing = 'outputing'
 hadler = 'hadler'
 aggragate = 'aggragate'
 qucik = 'qucik'
-alog = 'alog'
 exsit = 'exsit'
 deciamls = 'deciamls'
 uncorrectly = 'uncorrectly'
@@ -662,7 +653,6 @@ insid = 'insid'
 coodinate = 'coodinate'
 usefull = 'usefull'
 sqaure = 'sqaure'
-adn = 'adn'
 intialize = 'intialize'
 addtional = 'addtional'
 Taget = 'Taget'
@@ -705,7 +695,6 @@ defaut = 'defaut'
 formating = 'formating'
 infor = 'infor'
 becuase = 'becuase'
-annotaion = 'annotaion'
 temporaily = 'temporaily'
 defferent = 'defferent'
 Flattern = 'Flattern'
@@ -753,7 +742,6 @@ compitable = 'compitable'
 comple = 'comple'
 dealed = 'dealed'
 ser = 'ser'
-anc = 'anc'
 contraints = 'contraints'
 propogated = 'propogated'
 beacuse = 'beacuse'
@@ -777,7 +765,6 @@ craete = 'craete'
 expaned = 'expaned'
 choos = 'choos'
 whos = 'whos'
-architecuture = 'architecuture'
 argumet = 'argumet'
 coule = 'coule'
 instanciate = 'instanciate'
@@ -823,7 +810,6 @@ imformation = 'imformation'
 kernerl = 'kernerl'
 Boardcast = 'Boardcast'
 Greate = 'Greate'
-Alread = 'Alread'
 unkown = 'unkown'
 recevied = 'recevied'
 Normlized = 'Normlized'
@@ -832,7 +818,6 @@ orginal = 'orginal'
 Stati = 'Stati'
 Destory = 'Destory'
 seperately = 'seperately'
-alloctor = 'alloctor'
 fullfill = 'fullfill'
 Substitude = 'Substitude'
 producted = 'producted'

paddle/cinn/operator_fusion/graph_transformer/search_algorithm.h

@@ -138,8 +138,8 @@ struct SearchAlgorithm<ReverseTopoNodePairPattern,
 template <typename Kind, typename GraphMatcher, typename GraphOperation>
 void GraphTransformer(PatternGraph* graph) {
   VLOG(4) << "Start GraphTransformer...";
-  auto alog = SearchAlgorithm<Kind, GraphMatcher, GraphOperation>(graph);
-  alog();
+  auto along = SearchAlgorithm<Kind, GraphMatcher, GraphOperation>(graph);
+  along();
 }
 
 }  // namespace cinn::fusion

paddle/fluid/operators/fused/fused_attention_op.cc

@@ -389,8 +389,8 @@ class FusedAttentionOpMaker : public framework::OpProtoAndCheckerMaker {
                   "The qkv_w shape is (h, 3h), do transpose to it.")
         .SetDefault(false);
     AddAttr<bool>("pre_layer_norm",
-                  "if true, the attention op uses pre_layer_norm architecure, "
-                  "else, uses post_layer_norm architecuture. "
+                  "if true, the attention op uses pre_layer_norm architecture, "
+                  "else, uses post_layer_norm architecture. "
                   "[default false].")
         .SetDefault(false);
     AddAttr<float>("epsilon",

paddle/fluid/pybind/pybind.cc

@@ -2476,7 +2476,7 @@ All parameter, weight, gradient are variables in Paddle.
 
   m.def("_is_program_version_supported", IsProgramVersionSupported);
 #if defined(PADDLE_WITH_CUDA)
-  m.def("alloctor_dump", [](const phi::GPUPlace &place) {
+  m.def("allocator_dump", [](const phi::GPUPlace &place) {
     auto allocator = std::dynamic_pointer_cast<
         paddle::memory::allocation::AutoGrowthBestFitAllocator>(
         paddle::memory::allocation::AllocatorFacade::Instance()

paddle/phi/kernels/cpu/masked_select_grad_kernel.cc

@@ -41,16 +41,16 @@ void MaskedSelectGradKernel(const Context& dev_ctx,
 
   auto expanded_size = funcs::MatrixGetBroadcastBatchPortion(
       common::vectorize(x_grad->dims()), common::vectorize(mask.dims()));
-  auto expaned_dims = common::make_ddim(expanded_size);
+  auto expanded_dims = common::make_ddim(expanded_size);
 
-  if (mask.dims() != expaned_dims) {
+  if (mask.dims() != expanded_dims) {
     ExpandKernel<bool, Context>(
         dev_ctx, mask, IntArray(expanded_size), &mask_expand);
   } else {
     mask_expand = mask;
   }
 
-  if (x_grad->dims() != expaned_dims) {
+  if (x_grad->dims() != expanded_dims) {
     x_grad_expand = Empty<T, Context>(dev_ctx, IntArray(expanded_size));
     expand_x = true;
   } else {

paddle/phi/kernels/gpu/masked_select_grad_kernel.cu

@@ -66,16 +66,16 @@ void MaskedSelectGradKernel(const Context& dev_ctx,
 
   auto expanded_size = funcs::MatrixGetBroadcastBatchPortion(
       common::vectorize(x_grad->dims()), common::vectorize(mask.dims()));
-  auto expaned_dims = common::make_ddim(expanded_size);
+  auto expanded_dims = common::make_ddim(expanded_size);
 
-  if (mask.dims() != expaned_dims) {
+  if (mask.dims() != expanded_dims) {
     ExpandKernel<bool, Context>(
         dev_ctx, mask, IntArray(expanded_size), &mask_expand);
   } else {
     mask_expand = mask;
   }
 
-  if (x_grad->dims() != expaned_dims) {
+  if (x_grad->dims() != expanded_dims) {
     x_grad_expand = Empty<T, Context>(dev_ctx, IntArray(expanded_size));
     expand_x = true;
   } else {

python/paddle/distributed/passes/auto_parallel_fp16.py

@@ -124,7 +124,7 @@ def _keep_fp32_input(op, in_name):
 
 # TODO check if bf16 and fp16 still share the same logic
 def _keep_fp32_output(op, out_name):
-    # TODO(lizhiyu02): Support 'custom_white_list' adn 'custom_black_list' in amp_options
+    # TODO(lizhiyu02): Support 'custom_white_list' and 'custom_black_list' in amp_options
     if not op.amp_options.enable:
         return True
     op_type = op.type

python/paddle/incubate/asp/supported_layer_list.py

@@ -72,7 +72,7 @@ def _default_pruning(
     # In sparse training, layer weight matrices is viewed sparse matrix A, so
     # the math formula should be 'Act(WX + b)'. However, default formula in PaddlePaddle
     #  is 'Act(XW + b)'. For enabling SPMMA, weights and inputs should be transposed
-    # for computing, Act( (W^T X^T)^T + b). Therefore, we have to prune alog k dimension
+    # for computing, Act( (W^T X^T)^T + b). Therefore, we have to prune log k dimension
     # of W^T, which is m dimension of W. Moreover, all mask generating functions in
     # asp/utils is row-major pruning. That is the reason we have to transpose weight
     # matrices before invoking create_mask. Then we transpose the result mask to make

python/paddle/nn/utils/transform_parameters.py

@@ -33,7 +33,7 @@
     from paddle._typing import ShapeLike
 
 
-# input==output, inplace strategy of reshape has no cost almostly
+# input==output, inplace strategy of reshape has no cost almost
 def _inplace_reshape_dygraph(x: Tensor, shape: ShapeLike) -> None:
     x_shape = _create_tensor(dtype='int64')
     if in_dygraph_mode():

python/paddle/static/quantization/post_training_quantization.py

@@ -354,7 +354,7 @@ def __init__(
         self._quantized_weight_var_name = set()
         self._quantized_act_var_name = set()
         self._weight_op_pairs = {}
-        # The vars for alog = KL or hist
+        # The vars for along = KL or hist
         self._sampling_act_abs_min_max = {}
         self._sampling_act_histogram = {}
         self._sampling_data = {}

python/paddle/tensor/tensor.prototype.pyi

@@ -13,7 +13,7 @@
 # limitations under the License.
 
 # The `Tensor` template `tensor.prototype.pyi` for `tools/gen_tensor_stub.py` to generate the stub file `tensor.pyi`.
-# Add docstring, attributes, methods and alias with type annotaions for `Tensor` in `tensor.prototype.pyi`
+# Add docstring, attributes, methods and alias with type annotations for `Tensor` in `tensor.prototype.pyi`
 # if not conveniently coding in original place (like c++ source file).
 
 # Import common typings for generated methods

test/cpp/fluid/platform/errors_test.cc

@@ -92,7 +92,7 @@ TEST(Errors, NotFound) { CHECK_ALL_PADDLE_EXCEPTION_MACRO(NotFound); }
 
 TEST(Errors, OutOfRange) { CHECK_ALL_PADDLE_EXCEPTION_MACRO(OutOfRange); }
 
-TEST(Errors, AlreadExists) { CHECK_ALL_PADDLE_EXCEPTION_MACRO(AlreadyExists); }
+TEST(Errors, AlreadyExists) { CHECK_ALL_PADDLE_EXCEPTION_MACRO(AlreadyExists); }
 
 TEST(Errors, ResourceExhausted) {
   CHECK_ALL_PADDLE_EXCEPTION_MACRO(ResourceExhausted);

test/cpp/phi/kernels/test_gpu_timer.cu

@@ -86,19 +86,19 @@ TEST(GpuTimer, Sum) {
 #endif
 
   using Functor = std::function<void(float *, float *, size_t)>;
-  Functor alog0 = Algo<4, 256, 1024>;
-  Functor algo1 = Algo<1, 256, 1024>;
-  Functor alog2 = Algo<1, 256, 8>;
+  Functor along0 = Along<4, 256, 1024>;
+  Functor along1 = Along<1, 256, 1024>;
+  Functor along2 = Along<1, 256, 8>;
 
-  std::vector<Functor> algos = {alog0, algo1, alog2};
+  std::vector<Functor> algos = {along0, along1, along2};
 
   for (int j = 0; j < algos.size(); ++j) {
     auto algo = algos[j];
     phi::GpuTimer timer;
     timer.Start(0);
     algo(d_in1, d_in2, N);
     timer.Stop(0);
-    VLOG(3) << "alog: " << j << " cost: " << timer.ElapsedTime() << "ms";
+    VLOG(3) << "along: " << j << " cost: " << timer.ElapsedTime() << "ms";
   }
 
 #ifdef __HIPCC__

test/dataset/imikolov_test.py

@@ -31,7 +31,7 @@ def test_train(self):
         self.check_reader(paddle.dataset.imikolov.train(WORD_DICT, n), n)
 
         first_line = (
-            'aer banknote berlitz calloway centrust cluett fromstein '
+            'are banknote berlitz calloway centrust cluett fromstein '
             'gitano guterman hydro-quebec ipo kia memotec mlx nahb punts '
             'rake regatta rubens sim snack-food ssangyong swapo wachter'
         )

tools/gen_pybind11_stub.py

@@ -185,7 +185,7 @@ def wrapper(self, arg: Any):
 
 
 def _patch_pybind11_invalid_annotation():
-    # patch invalid annotaion as `Value`, e.g. 'capsule' to 'typing_extensions.CapsuleType'
+    # patch invalid annotation as `Value`, e.g. 'capsule' to 'typing_extensions.CapsuleType'
     def wrap_name(func):
         @functools.wraps(func)
         def wrapper(self, arg: Annotation):